PROJECT SUMMARY The ultimate vision of this proposal is to develop a technology platform for the direct, rapid, and robust detection of cancer-associated DNA mutations for diagnostics and research. DNA mutations have been known to be fundamental to cancer development for decades and specific fragments of mutated DNA have been sought as non-invasive biomarkers for cancer in blood, stool and other samples. However, a major barrier to progress has been the lack of highly sensitive, specific, and quantitative methods for detecting DNA mutation when mutant DNA fragments are present as rare alleles in a high background of wild-type DNA. We propose here an approach to DNA mutation detection and quantification that is conceptually simple, yet takes advantage of sophisticated and elegant advances in single molecule imaging science. The approach is based on using total internal reflection microscopy to detect the binding and release of a fluorescently-tagged probe to immobilized target DNA molecules on the surface of a glass slide. Differences in nucleotide sequence even at a single nucleotide, give rise to differences in the free energy of hybridization, and this affects the kinetics of binding and release of a probe in a manner that can be detected by single-molecule microscopy. We have already performed proof-of-concept and established feasibility of the approach. In this project, we aim to perform optimizations that will increase sensitivity and increase throughput, and to perform validation with clinical samples from lung cancer patients and controls.